584,826 active members*
4,985 visitors online*
Register for free
Login
Page 3 of 4 1234
Results 41 to 60 of 70
  1. #41
    Join Date
    Jul 2018
    Posts
    6248

    Re: 150x100cm Steel frame router build

    Hi Niko - singular: axis plural: axes advice: singular advice: plural my wife's an english teacher and I regularly get beaten up on my english...Peter advice is an uncountable noun meaning single and plural are the same (Just looked it up)

  2. #42
    Join Date
    May 2011
    Posts
    790

    Re: 150x100cm Steel frame router build

    Quote Originally Posted by Nikolaguca View Post
    I've bought the tube today and I now realize what you are talking about... a 135cm piece of tubing is 220lbs/100kg
    Yep, and do you have a drill press / mill to drill holes? You might not be able to get it to fit under there with a bit. You might have to hand drill it.

    You don't have to commit to it, you know if you sold it back for scrap, you might only loose $40. Sounds like there are a couple loose ends to the design still.

    Quote Originally Posted by Nikolaguca View Post
    I haven't bought the motors yet. I thought about putting a 2:1 reduction on a 8.5Nm Nema 34 since they are terrible at speeds close to 1000rpm
    What's the opposite of gear reduction called? Gear increasion? LOL. I think that's what you're talking about. No, I was talking about using a servo with gear reduction so that the servo spins faster than the ballscrew. For the stepper, instead of gear increasion (I made that word up), you may consider a higher lead of ballscrew. You haven't bought the motors? Great, don't. Just hold off. You may go with the JMC servos for about the same cost.

    Quote Originally Posted by Nikolaguca View Post
    I've found a 3KW spindle and VFD for $250 so that's what I'll use for now... if I decide to mill aluminium I'll buy BT30 spindle for that.
    I have a monster spindle that weighs 100 lbs and I still haven't got it running yet. I've come to realize it might be better to user a 6mm single flute bit running at higher RPMs, with a decent acceleration and speed making light cuts and adaptive toolpaths, you can get decent material removal rates.

    What's the weight of your spindle system? You might need a pneumatic counterbalance, be it a low damping gas strut, or two, or something else.

    Quote Originally Posted by Nikolaguca View Post
    Since I'll drive the X axis with a driven ballnut .
    That's a whole level of complicated that I can't help you with. It changes everything in regards to the math too. You're complicating things alot here. Are you calling your gantry the X axis?

    Quote Originally Posted by Nikolaguca View Post
    I can have multiple Z axis
    My understanding is that people often use the multiple Z axis on routers so they don't have to change bits. To a lesser extent to have the option of a spindle with more torque at lower speed and a higher speed spindle on the same machine. If your two Z axis move independently from each other, then they need to be homed independently and reliably if you were to use both on the same job. This adds complexity to the electronics and your board would need to handle that. I'm not an expert on this subject.

    Did you read the Linux_Fan thread to see what he did there?

    Quote Originally Posted by Nikolaguca View Post
    I'm sorry now that I had to go away from a high rail design since that would be much more rigid than this setup...
    At the risk of sounding repetative, what you have shown as built would work with a similar design to Linux_Fan's. And his kicks butt. But the choice is always yours to make. The easy thing about his design is that the aluminum plate should come flat enough for your purposes, and it's easier to work with. I won't talk about it anymore.

    Quote Originally Posted by Nikolaguca View Post
    Is the red line on the graph for your motor? Or do the different lines represent different operating voltages for the same motor? If so, what voltage are you going for?

    What is the rotor inertia for these motors? Should be in the spec sheet.

  3. #43
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    I've welded the construction today... just to figure out how to get it off my welding table and I'll send it to machining. It's around 200kg already.

    I didn't weld it fully and I did jump all over the place so the frame didn't heat up much at all. I'll fill the gaps in the weld with bondo so I can fill the tubes with sand later.

    I actually thought about 2:1 reduction on the Y axis where I'll have two 8.5Nm motors - motors spinning at 500rpm and the screw at 1000rpm. Or at least that's what I thought about doing with a lighter gantry. I haven't done any calculations yet so I'll do that down the road.

    Servos are out of my budget right now. These 8.5Nm steppers are $150 and the cheapest servos I could find are over $250 for a 400w one.

    How do you compare the power of servo vs stepper motor? Is 400w servo more powerfull than a 8.5 stepper?

    Yes, the red line on that chart is for my motor. That's why I wanted to reduce it 2:1.

    Gosh! So many questions. One more; Would filling Y axis beams with concrete stiffen them up by any significant amount? I would add those glass fibers to the concrete as well. Or should I just stick with sand and cry because I didn't use heavier tube? The tube is 120x80x4mm with 6mm flat bar welded on top.

  4. #44
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    Next weekend I'll send the frame for machining and that leaves me to start working on the gantry.

    The main concern I have right now is about how to connect it to the carriages below the table.

    My aim is to use thick walled 3x3 steel tubing and to either have 1/2" angle iron connecting them to the carriages or some 1/2" flat plate. I would add some bracing (not a lot since I don't have a lot of space).

    In my head this is a weakest link in this design but what can I do right now... I must stick with it. But with 1/2" thick material I think this'll be rigid enough.

    Do you have any thoughts on this? Thnaks.

  5. #45
    Join Date
    Jan 2008
    Posts
    1523

    Re: 150x100cm Steel frame router build

    I'm not sure why you went with that design? To protect the rails from dust and chips? Linear bearings have good seals and building a weaker design just to cover them is a bad idea. Dust and chips should be stopped at the source (dust collection), or a simple bellows cover over the rails.

    I guess it is too late now to change your design?
    7xCNC.com - CNC info for the minilathe (7x10, 7x12, 7x14, 7x16)

  6. #46
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    Quote Originally Posted by pippin88 View Post
    I'm not sure why you went with that design? To protect the rails from dust and chips? Linear bearings have good seals and building a weaker design just to cover them is a bad idea. Dust and chips should be stopped at the source (dust collection), or a simple bellows cover over the rails.

    I guess it is too late now to change your design?
    I went with this design because the gantry mill that will machine this frame is 100cm wide, and I had to put the rails below the table if I'm going to have a 100cm wide machine.

    Putting the rails on the sides would be a good idea but they couldn't guarantee me that the sides would be 100% parallel, they could get out of parallel by 0.3-0.4mm over 150cm.

    Sent from my SM-G970F using Tapatalk

  7. #47
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    With the main frame fully welded I can now start working on the X axis. Since I'm not sure what I'll use the machine for, I'm not sure at what height to set the gantry at. I'm thinking 9-10 inches will be enough.

    @NIC 77 You've tickled my mind with those JMC servos... I've been googling "stepper vs servo" for days now and I'm close to decide to go with servos. Even though the cheapest ones are still twice as expensive as steppers, when it comes to power and speed, there is no contest.

    But I'm concerned with tuning. DMM and Delta offer some auto-tuning software but they are way out of my price range. Can a servo motor run without tuning? Is manual tuning possible for a DIY guy like me?

    Something like this would be ~$350-$400 per motor:
    https://www.ebay.com/itm/400w-3000rp...EAAOSwTvNd5Mp4

    Do I need brake on any other axis apart from Z? Also can you have both steppers and AC servos on one machine? I would put stepper on Z axis since they are cheaper...

  8. #48
    Join Date
    Jan 2005
    Posts
    15362

    Re: 150x100cm Steel frame router build

    Quote Originally Posted by Nikolaguca View Post
    With the main frame fully welded I can now start working on the X axis. Since I'm not sure what I'll use the machine for, I'm not sure at what height to set the gantry at. I'm thinking 9-10 inches will be enough.

    @NIC 77 You've tickled my mind with those JMC servos... I've been googling "stepper vs servo" for days now and I'm close to decide to go with servos. Even though the cheapest ones are still twice as expensive as steppers, when it comes to power and speed, there is no contest.

    But I'm concerned with tuning. DMM and Delta offer some auto-tuning software but they are way out of my price range. Can a servo motor run without tuning? Is manual tuning possible for a DIY guy like me?

    Something like this would be ~$350-$400 per motor:
    https://www.ebay.com/itm/400w-3000rp...EAAOSwTvNd5Mp4

    Do I need brake on any other axis apart from Z? Also can you have both steppers and AC servos on one machine? I would put stepper on Z axis since they are cheaper...
    With Dmm there is basically no tuning required just a basic setup and run , only the Z axis may need a brake, not all need a brake even on the Z axis unless it is very heavy

    And 400w are not going to run very well on your machine unless you are gearing them down
    Mactec54

  9. #49
    Join Date
    Jan 2008
    Posts
    1523

    Re: 150x100cm Steel frame router build

    I haven't seen a JMC manual yet. Generic Chinese servos (sold under many brands) have difficult to understand manuals.

    Delta have a good English manual. Not a poor translation.

    Servos are not necessarily plug and play and so to me a decent manual (and maybe tuning software) is important.

    You can size your servos with software. https://motioneering.kollmorgen.com

    Requires an account and a little bit of learning.
    7xCNC.com - CNC info for the minilathe (7x10, 7x12, 7x14, 7x16)

  10. #50
    Join Date
    May 2011
    Posts
    790

    Re: 150x100cm Steel frame router build

    Quote Originally Posted by Nikolaguca View Post
    With the main frame fully welded I can now start working on the X axis. Since I'm not sure what I'll use the machine for, I'm not sure at what height to set the gantry at. I'm thinking 9-10 inches will be enough.
    That's always the big question. How much Z travel. Too much Z travel, and your Z axis needs to become very stiff. I'm working on a dual column, rising gantry, moving table design. This solves the problem but limits the work area and adds extra complexity like counterbalancing the gantry.

    The first machine I made had two positions for the table. I never once used the lower position.

    Start by figuring out what kind of work holding, such as vices, etc. you might want to use.

    Quote Originally Posted by Nikolaguca View Post
    @NIC 77 You've tickled my mind with those JMC servos... I've been googling "stepper vs servo" for days now and I'm close to decide to go with servos. Even though the cheapest ones are still twice as expensive as steppers, when it comes to power and speed, there is no contest.
    I've actually been looking at the integrated servos from JMC for my industrial 3d printer build. After some back and forth conversations with them on Alibaba I have discovered that they have 1000 line incremental encoders with quadrature decoding giving 4000 CPR (counts per rev). This is the same as most of the Leadshine closed loop stepper as far as resolution goes.

    I've never used them obviously, so I can't give you a review, but they seem quite popular with the European CNC building crowd. I'm not sure if I like the integrated drivers from a perspective of vibration isolation of the drivers. Also, I assume that you plan on a 240V single phase power supply for this machine? You'll need to buy a separate DC supply for the integrated versions.

    They have 400W, 440W, and 660W versions.

    I'm considering 4 of the 400W servos for myself.

    https://www.alibaba.com/product-deta...21160185.html?

    https://www.alibaba.com/product-deta...69541028.html?

    For a 400W or 440W, off the top of my head, a 10mm lead ballscrew with 3:1 belt driven gear reduction seems like it could be good.

    For direct drive you would be looking at the 660W or something like that, and you would never get the full potential out of the servo, but it might be worth it to make the construction easier. It's really your call.

    I can look at the numbers at some point, if you want, just really busy for the next few days. What are the total lengths of all of your ballscrews again?

    For my purposes, the cost of the 400W integrated servos on Alibaba is really close to a stepper and good driver, or some used AC servo drivers to power the servos I already have (which may or may not work). Something else may be more appropriate for you, I don't know. But these do have some good reviews.

    Quote Originally Posted by Nikolaguca View Post
    But I'm concerned with tuning. DMM and Delta offer some auto-tuning software but they are way out of my price range. Can a servo motor run without tuning? Is manual tuning possible for a DIY guy like me?
    JMC has it's own tuning software that you can download. The big question I have is how to tune both servos on a dual driven gantry. I think you might need to rig something up to simulate the weight and tune each one individually like that. Not sure. I'm not really an expert on this. But really, sure you can figure it out. I doubt it's going to be that hard.

    One comment I copied from a Youtube video (can't seem to find that video again) about the integrated ones was:

    Quote Originally Posted by Random YouTuber
    Yeah, these servo's are amazing value for their money! I'm impressed by them as well. (using 4 of them on my machine) Make sure to get the V6 instead of the V5's if you don't like messing with the PID settings. The V6 has an auto-tune function.
    Quote Originally Posted by Nikolaguca View Post
    Do I need brake on any other axis apart from Z? Also can you have both steppers and AC servos on one machine? I would put stepper on Z axis since they are cheaper...
    You can have both on one machine. I know with the Duet board I plan on using, you need to adjust the microstep parameters to get the servos to work and other people have been doing that with good success. For each specific board and control software there's going to be different things to consider. I'm not the guy to ask about specifics on each set up.

    Here's a thread on that:

    https://forum.duet3d.com/topic/12126...xis-movement/5

    How heavy is your Z axis going to be? You might want a pneumatic counterbalance regardless. That could be as simple as two gas struts. Ones without much damping would be preferable.

    Here's a thread you might find interesting:

    https://www.usinages.com/threads/cnc...uminium.63531/

  11. #51
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    On some machines I saw that people put 3 bearing blocks on each side of the long axis...

    Is there any benefit in doing so regarding rigidity and life span on the bearing blocks?

    I'm using 25 HGR linear bearings and rails. I would add a third block if that would increase rigidity in any noticable way... they are relatively cheap nowdays anyway

    Sent from my SM-G970F using Tapatalk

  12. #52
    Join Date
    Jul 2018
    Posts
    6248

    Re: 150x100cm Steel frame router build

    Hi Niko - The middle car would be doing nothing. If you have not bought the cars yet the best plan to increase rigidity and life is to use heavy pre-load bearings. These are considerably stiffer then "normal" spec and as the preload stops internal movement in the car at changes of direction so life improves. The only way the middle car would be doing anything is if the dead load was near the capacity of the cars. 25mm cars carry at least Co= 34kN say 34000/10=3400kg so does your machine weight 6.8 tonne per side? Regards Peter

  13. #53
    Join Date
    May 2011
    Posts
    790

    Re: 150x100cm Steel frame router build

    Peter,

    I've heard you say before that you should go with heavily preloaded cars. And that has me wondering.

    Are you talking about the inexpensive cars that you can get from China or is there a particular brand name that you are referring to? Can you even order the generic Chinese linear rails in different preloads?

    From the video reviews I've seen of these rails there is so much hit and miss on the manufacturing end that what you get regarding any preload or slop is a roll of the dice.

    What has your experience been with the different rails that you have bought for your projects? What kind did you buy, brand name, etc? Do you have a good supplier that you can recommend?

    On a DIY machine where alignment might not be perfect, and there are several cars per axis, I'm not convinced that the heaviest preload is the best option.

    If you look at TBI ballscrews as an example, they don't recommend their highest preload to be used for CNC machines.

    The Bosch Rexroth and NSK rails I have are awesome. To me, brand name is important for linear rails. I used THK rails at work, and they were OK, but required some adjustment, alot of lubrication and a break in period before they ran smooth without the gritty start-stop feeling (like driving over a washboard road). Still, once they were broken in and lubricated they were fine, and still a huge step up from generic rails and cars.

    I'd rather go with surplus Rexroth carriages, made in Germany, with light or no preload than anything made in China. The quality of these products is just amazing. Believe me when I say that you can not feel any slop whatsoever in the zero preload Rexroth cars. They are very very tight! And smooth! But you have to find them as new or used surplus to be able to afford them.

    The pics below come from the Rexroth ball rail systems catalog that you can download from their website.

    Attachment 449090

    I'd like to point out what it says there: "this clearance is usually equalized by parallelism tolerances".

    Attachment 449092

    If you look at the C3 preload for size 25, 2770N Pre-tensioning force, that's 622 lbs of pre-tensioning force on each car! What does pre-tensioning force mean exactly?

    Attachment 449094

    You can see that the different preloads, C1, C2, C2, you really have to put alot of force on their to get deflection, and that the difference between the 3 preloads isn't that big. This is for size 35, didn't see this chart for size 25 in the catalog, but I'm guessing it's somewhat similar.

    Keep in mind, I'm not saying that you're wrong, I'm just saying that I'm not convinced going with the highest preload is the best option on a DIY machine build, and also I'm definitely saying that IMO brand name is really really important when it comes to linear rails and CNC. I'm just opening it up for discussion, something I may be able to learn from as well.

  14. #54
    Join Date
    Jul 2018
    Posts
    6248

    Re: 150x100cm Steel frame router build

    Hi Nic - Brand is very important as this is related to quality and $$$. I have used WON, STAF and PMI brands. When new you cannot feel any clearance on zero or light preload cars. The main sticky bit is the seal. My experience is that after 250hrs work the clearance drops one level. ie if you have light preload they become zero, if you have zero they become clearance etc. After 500hrs I can clearly feel the clearance. I have not run a heavy for 500hrs yet so can't report on that. The better manufactures produce detailed design manuals. If you are up to the math you follow their recommendations. I'm about to get linear rails and cars from BST their house brand in heavy. Same cost just more lead time as they are not stocked. The way this works is as follows. All the rails are made to the same tolerance within the tolerance band. The bearings are made in different sizes so that's the difference, higher preload uses bigger balls. You can buy the balls separate and reload them as well. eg if you have a worn car you get the next size up balls and reload.

    Take the 622lbs of preload. You can't feel this as its an internal in equilibrium load. Maybe the friction is slightly up but not by much. This means when a load changes direction the cars motion does not suffer hysteresis or the balls don't bounce (unless the applied load goes over the preload). I've attached the Hiwin 20mm car stiffness chart. You can see that the zero clearance cars deform more then the high preload cars. Think of preload as a bolted connection. If you torque the bolt to say 500kgf then if you pull on the bolt with 501kgf you will gap the connection. In the case of a car you don't want the car to gap as this will create a jerk and that jerk will damage the balls (brinelling) and eventually they will fail (fatigue).

    When I buy cars they are smooth from the get go. Hope this helps Regards Peter

  15. #55
    Join Date
    Jul 2018
    Posts
    6248

    Re: 150x100cm Steel frame router build

    Hi Nik - re parallelism tolerance. This implies the 1-10um clearance is taken up by the rail spacing ie the rails are set out (or in or wavy) further then the theoretical center distance. This means the car is stiff in one direction but in the other direction it has a 10um gap. If set up at the correct centreline distance you would have a 5um gap both ways on average or away from the load side a 10um gap.. And that 10um may sound small but its a long way to the tool so that 10um will get magnified somewhat at the tool.

    This is not good in router and mill applications as it creates jerks and inaccurate paths when the load direction changes. In non contact applications this is acceptable for slow speeds but for high speeds inertial loads may jerk the car in the gap direction which is something you may not want. If your happy with +/-0.1mm then 10um does not matter (10um =0.010mm) But if your aiming at +/-0.010mm or +/-0.001mm then it does...So its horses for courses and you get what you paid for. You have to understand the classes of cars and your application and costs to decide what you need. For routers and mills use the highest preload you can get at the price your happy to pay. Regards Peter

  16. #56
    Join Date
    May 2011
    Posts
    790

    Re: 150x100cm Steel frame router build

    Quote Originally Posted by peteeng View Post
    Hi Nic - Brand is very important as this is related to quality and $$$. I have used WON, STAF and PMI brands.
    https://www.wonststore.com/Linear%20Motion
    Professional Linear Motion Solution Provider in Taiwan. - ??????????
    PMI-GROUP// PMI PRECISION MOTION INDUSTRIES, INC.

    WON is from Korea
    STAF is headquartered in Taiwan, and it looks like they also have China manufacturing
    PMI is from Taiwan

    Yeah, those look really decent, definitely not the typical discount bearings people are buying on eBay. That gives a few more options to people looking for bearings.

    Quote Originally Posted by peteeng View Post
    My experience is that after 250hrs work the clearance drops one level. ie if you have light preload they become zero, if you have zero they become clearance etc. After 500hrs I can clearly feel the clearance.
    I didn't anticipate the bearings wearing out like that.

    Quote Originally Posted by peteeng View Post
    I'm about to get linear rails and cars from BST their house brand in heavy. Same cost just more lead time as they are not stocked.
    I'm looking forward to hearing a review on them. I did not know you could get them in heavy preload. That answers that question, but I wonder what they will be like. There are so many different products from China, it would be nice if they had more brand names to differentiate between them.

    Quote Originally Posted by peteeng View Post
    When I buy cars they are smooth from the get go.
    It looks like you've been buying decent stuff. Korean and Taiwanese products. I'm looking forward to hearing what you think of the BST rails and cars. Is that specific "brand" only available from BST? Do they manufacture them? I like BST, they have good customer service, I'm just curious what you're actually buying.

    It would be nice if you could also get your hands on some more generic Chinese ones for a comparison. If I'm not mistaken that is what half of the people who are making machines here are buying. In the reviews I've seen the tolerances can be very hit and miss, and every once in a while someone makes a post about a machine bought from China that came with a bad bearing or two.

  17. #57
    Join Date
    Jul 2018
    Posts
    6248

    Re: 150x100cm Steel frame router build

    Hi Nic77 - My machines are designed as production machines so I can't just pick something from ebay or amazon etc. They have to be repeatable and consistent. For one off machines you can shop and get things at less price but that price means they may not be stocked again or they are low cost due to low "quality". I quote quality as quality is proportional to cost and everyone has a budget & a purpose. I was getting good low cost bearings from a local australian company but they decided not to stock that line at some point. That mucked up my costing and supply line. Then I had to find an asian supplier and BST was mentioned in this forum. Others are pleased with BST and so far Fred has answered all my Q's promptly. I do ask a lot of Q's!! cheers Peter

    The house brand BST stuff is marked BST. So you will need to ask them. They also stock Hiwin. I don't think the bearings "wear out" they are rated at 50,000 kms lifetime. I think they just settle into their true tolerance band. Initially I bought std shelf stock from a company and they turned out to be zero clearance. On a rebuild I noticed them to be gappy so investigated further. Thats how you learn about these things not from the books. So I then replaced these with high preload and the machine definitely less wobbly at tight corners. Since UCCNC (see statistics page) has a usage clock on it you can see how far each axis has travelled and how much time is on it. All good info. Peter

  18. #58
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    Can you re-ball the bearing blocks with bigger balls to get a higher preload just as you do with ballscrew nuts? Is it simple as that or not?



    Sent from my SM-G970F using Tapatalk

  19. #59
    Join Date
    Jul 2018
    Posts
    6248

    Re: 150x100cm Steel frame router build

    Hi Niko - I have not tried it but the bearing people sell the balls in graded sizes so it "should" be simple, tell us how you go Peter

    OPPs I said the bearings where rated at 50,000kms I think it maybe 50kms will need to reread the data sheets...Yes I meant 50,000m or 50km heres the extract from the Hiwin manual.. Peter
    Attached Thumbnails Attached Thumbnails life.JPG  

  20. #60
    Join Date
    Jan 2020
    Posts
    39

    Re: 150x100cm Steel frame router build

    Isn't that 50km figure if the bearings operate at their maximum dynamic load?

    At 10m/min you would go through that 50km in 5 days...


    Sent from my SM-G970F using Tapatalk

Page 3 of 4 1234

Similar Threads

  1. Welded steel frame router build
    By jones in forum DIY CNC Router Table Machines
    Replies: 94
    Last Post: 09-06-2023, 09:57 AM
  2. Steel Frame Router Build
    By trailerparkboys in forum CNC Wood Router Project Log
    Replies: 1
    Last Post: 03-08-2016, 01:53 PM
  3. Steel frame Ganrty router build
    By matth in forum CNC Wood Router Project Log
    Replies: 172
    Last Post: 10-30-2014, 08:32 PM
  4. linear bearings and steel frame considerations on new build
    By Thermwood in forum Linear and Rotary Motion
    Replies: 12
    Last Post: 11-08-2013, 07:14 PM
  5. My steel frame router
    By Spiv in forum DIY CNC Router Table Machines
    Replies: 18
    Last Post: 06-19-2009, 12:14 AM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •